~__ JOURNALOF

Fmancial
ELSEVIER Journal of Financial Economics 46 (1997) 3-28
ECONOMICS

Why is there a home bias?

An analysis of foreign portfolio equity ownership
in Japan ~
Jun-Koo F a n g ~, Ren6 M. Stulz *'b
"College of Business Administration, Korea UniversiO'. 5-1 Anam-Dong, Sung B u k - G U. Seou1136- 701,
South Korea
hDepartment of Finance, Max M. Fisher College of Business, ] 775 College Road, Ohio State Universi~,.
Columbus, OH 43210, USA

Received 29 June 1995; received in revised form 23 January 1997

Abstract

This paper studies stock ownership in Japanese firms by non-Japanese investors from
1975 to 1991. Existing models predicting that foreign investors hold national market
portfolios or portfolios tilted towards stocks with high expected returns are inconsistent
with our evidence. We document that foreign investors hold disproportionately more
shares of firms in manufacturing indnstries, large firms, and firms with good accounting
performance, low unsystematic risk, and low leverage. Comrolling for size, there is
evidence that small firms that export more, firms with greater share turnover, and firms
that have A D R s have greater foreign ownership.

Keywords." Japan; H o m e bias; International diversification; Equity ownership

JEL classification. G 11; G15

* Corresponding author.
1 Some of the work on this paper was done while Stulz was a Bower Fellow at the Harvard
Business School. We are grateful for comments from Marianne Baxter, Eric Falkenstein, Wayne
Ferson, Stuart Gilson, Yasushi Hamao, David Mayers, Stewart Myers, Tim Opler, Scott Mason,
Jim Poterba, Linda Tesar, Robert Shiller, Dick Thaler, Raman Uppal, Maria Vassalou, Jerry
Warner, an anonymous referee, and participants at the February 1995 Meeting of the NBER
Behavioral Finance Group, the 1995 NBER Summer Institute, the NYSE Conference on New
Developments in International Finance, the Columbia University Conference on Japanese Finance,
the American Finance Association meetings in San Francisco, the Georgia Tech International
Finance Conference, seminars at Harvard University and the Ohio State University, and for useful
conversations with Ingrid Werner.

0304-405X/97/S17.00 .,() 1997 Elsevier Science S.A. All rights reserved

PII S 0 3 0 4 - 4 0 5 X ( 9 7 ) 0 0 0 2 3 - 8
4 J.-K. Kang, R.M. Stu&/Journal of Financial Economics 46 (1997) 3 28

1. Introduction

Over the last 25 yeaes, we have seen a dramatic decrease in obstacles to

international portfolio itavestment. Before the 1970s, most countries had restric-
tions on foreign exchange transactions t]hat limited cross-border portfolio
investment. Few developed countries have such restrictions now and many
emerging countries are eliminating them. Of Len, countries would forbid inbound
foreign portfolio investment altogether. Such a practice has almost disappeared,
although many countries still have limitations on foreign ownership. In
addition, the lack of tax harmonization meant that foreign investors often
would find it difficult to get refunds or credits for taxes paid abroad. This is
less of a problem now. Although not all obstacles have disappeared,
investors in most countries can now invest abroad both directly and through
mutual funds.
Financial economists have noticed that even though the barriers to interna-
tional investment have fallen dramatically, foreign ownership of shares is still
extremely limited and much smaller than one would expect in the absence of
barriers to international investment. French and Poterba (1991), Cooper and
Kaplanis (1994), and Tesar and Werner (19'95) document this phenomenon. In
particular, if investors care only about the mean and the variance of the real
return of their invested wealth, and if barriers to international investment are as
small as many observers suggest, one would expect investors, as a first approxi-
mation, to hold the world market portfolio of stocks. The available data on
ownership of shares shows that shares are mostly held by domestic residents, at
least for large countries such as the US and Japan. Several explanations have
been suggested for this so-called home bias in portfolios, but so far no explana-
tion seems to be generally accepted.
Existing investigations of the home bias; use countrywide data rather than
firm-specific data. In other words, we know well that non-Japanese investors
hold too few Japanese shares, but we do not know how foreign holdings in
Japan are divided among shares. In this study, we provide new insights into tile
home-bias puzzle by using disaggregated data for Japan. Although foreign
investment in Japan is interesting on its own merits, Japan also happens 1:o
be the only large country that we know for which detailed data on holdings by
foreign investors are available. In many countries, there are shares that
cannot be bought by foreign investors and others that can only be bought
by foreign investors (see Stulz and Wasserfallen, 1995). For these countries,
therefore, one can estimate shares held by foreign investors, but these
estimates reflect binding constraints on foreign ownership. In the case of Japan,
the foreign ownership constraints are not binding and therefore the data
on foreign ownership reflect the choices of foreign investors. We use a dataset
that provides foreign ownership for individual firms in Japan for a period of 17
years.
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3-28 5

Our most robust result is that foreign investors invest primarily in large
firms: on average, they hold 6.97% of the equity of the firms in the top
size quintile, compared to 1.21% of the equity of firms in the smallest size
quintile. Throughout our sample period, foreign investors in Japan have dispro-
portionately high holdings of firms in manufacturing industries and firms with
good accounting performance, h)w leverage, high market-to-book ratios, and
with low unsystematic risk relative to the weights of the Japanese market
portfolio.
Why do foreign investors prefer large firms? Firm size iLscorrelated with many
firm attributes that might affect investors' portfolio holdings. We therefore
investigate what makes large firms more attractive. First, it could be that
large firms are better known internationally. For instance, large firms are
more likely to sell goods abroad. We find some evidence that smaller firms that
export more have greater foreign ownership. Second, large firms could be more
liquid, so that foreign investors would find it cheaper to establish a position
without inside knowledge of the market. Consistent with this view, we find
evidence that the firms with the lowest turnover (defined as volume divided by
the number of shares outstanding) have less foreign ownership. Finally, inves-
tors in large firms could face fewer barriers to international investment. For
instance, the shares of these firms might be more easily available outside Japan.
We find that firms with American Depository Receipts (ADRs) have more
foreign ownership, although this was true even before the ADR program was
established.
The home-bias literature has emphasized that investors do not hold the world
market portfolio. In this paper, we show that, in addition, when investors invest
outside their home country, they do not hold the market portfolio of the
countries in which they invest. The evidence presented in this paper is inconsist-
ent with simple explanations of the home bias that assume that foreign investors
face similar obstacles to holding all shares in a country. The obstacles to foreign
investment must be inversely related to size to explain non-Japanese ownership
of shares in Japanese firms. However, foreign ownership, is small even for large
firms relative to what it would be if foreign investors l~eld the world market
portfolio.
The paper proceeds as follows. In Section 2, we review the implications of the
literature on international portfolio choice for equity portfolios of foreign
investors within a country. In Section 3, we describe our data and provide
annual summary measures of foreign investment in Japan for our sample period.
Section 4 investigates the determinants of foreign ownership using multivariate
regressions. In Section 5, we pursue several possible explanations for the size
bias we observe. In Section 6, we evaluate the performance of foreign investors
in Japan relative to the performance of the market. In Section 7, we analyze the
relation between foreign ownership and the cross-sectional variation in returns.
A conclusion is provided in Section 8.
6 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

2. The firm-level implications of possible explanations for the home-bias puzzle

M a n y papers have pointed out that despite the decline in barriers to interna-
tional investment, investors allocate only a very small fraction of their portfolio
to foreign investments, z This evidence constitutes what is generally referred to as
the home-bias puzzle. A number of explanations have been proposed for this
puzzle, but none individually has succeeded in resolving it. Each of these
explanations has implications for the holdings of foreign stocks by domestic
investors not only at the country level but also at the firm level. The most
important explanations for the home-bias puzzle and their implications f~r
foreign investment at the firm level involve explicit and implicit barriers
to international investment as well as departures from mean-variance opti-
mization.
Explicit barriers to international investment are those that are directly ob-
servable and quantifiable. For instance, a restriction on foreign exchange trans-
actions is an explicit barrier to international investment. Explicit barriers to
international investment have fallen over time because of, for instance, interna-
tional tax accords and the removal of foreign exchange controls. However, there
are still visible barriers to foreign investment, so that some home bias should still
exist. French and Poterba (1991) and Cooper and Kaplanis (1994) argue that
explicit barriers to international investment are no longer large enough to
explain the observed portfolio allocations of investors. They suggest that, to
explain the home-bias puzzle, these barriers would have to be much larger than
withholding taxes, which often are mentioned as the most significant observable
deterrent to foreign investment. If barriers to international investment are the
same across securities in a foreign country, foreign investors should tilt their
portfolios toward securities that have a higher expected excess return (see Stu].z,
1981a). If explicit barriers differ across securities, foreign investors will prefer
securities that have lower explicit barrier,;. We investigate this by examining
separately the ownership of firms with ADR programs, since these programs
should decrease transaction costs and holding costs for foreign investors.
Implicit barriers to international investment, on the other hand, are not
directly observable. As explicit barriers have fallen, researchers have put more
emphasis on obstacles to foreign investment that cannot be identified from
brokerage statements. The two main classes of such barriers are political risk
differences between domestic and foreign investors and information asyra-
metries.
Political risk differences arise if non-resident investors feel that there is some
probability that they might have trouble repatriating their holdings or that their

2 See Frankel (1995) for references as well as French and Poterba (1991L Cooper and Kaplanis
(1994I, and Tesar and Werner (1995).
 J.-K. Kang. R.M. Stu&/Journal of Financial Economics 46 (1997) 3 28 7

holdings might be expropriated altogether, so that their expected return on

foreign shares is lower than the expected return for residents. As long as this
political risk does not materialize, investors appear to be insufficiently diversi-
fied internationally. A problem with this argument is that money markets seem
well integrated at short maturities and do not reflect potential political risks.
Hence, to make the political risk argument convincing, one has to explain why
foreign investors would be more at risk with equities; than with short-term
money market instruments. One possibility is that short-term money market
instruments are highly liquid, so that investors can change their positions
quickly and at low cost. In contrast, markets for stocks, are less liquid, so that
investors may find it expensive to sell stocks quickly to avoid political risk. The
possibility of unexpected surges in political risk would suggest that foreign
investors invest more in securities that have liquid markets. We explore this
hypothesis with a study of the relation between foreign ownership and share
turnover as a proxy for liquidity.
As for information asymmetries, if non-resident investors are less well in-
formed about a country than resident investors, they will invest less in that
country because the variance of their predictive distribution is higher. 3 One can
think of some information asymmetries that affect all :securities similarly and
others that do not. Information asymmetries that affect all securities in a similar
way have no impact on portfolio allocation within a country. Merton (1987)
argues that investors invest in the securities they know about. If investors
behave this way, we would expect foreign investors to invest more in securities
that are known abroad. Since heavy exporters are presumably better known
abroad, the extent to which a firm exports may be a good proxy for how well it is
known abroad. We therefore examine the Merton hypothesis by investigating
whether foreign ownership is higher in firms that export more. In addition to
having different information, investors could process information differently
because of cognitive biases. French and Poterba (1990) argue that the holdings
of Japanese investors in the US and the holdings of US investors in Japan can be
explained if Japanese investors are substantially more optimistic about the
expected return of Japanese shares than are American investors. Shiller et al.
(1990) provide some survey evidence consistent with the view that investors are
more optimistic about their own market than are foreign investors.
Finally, departures from mean-variance optimization would suggest that
investors might be tailoring their asset holdings to hedge against changes in
variables that matter to them. Stulz (198la) .argues that investors' desire to

3See Low (1993) for a model where investors are asymmetrically informed and a home bias
emerges. Low's model assumes that the investmentopportunity set changes over time. Brennan and
Cao (1996) build a model where asymmetricinformation leads investors to buy foreign assets when
their return is high and sell them when their return is low.
8 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

hedge against unanticipated changes in their consumption and investment

opportunities might lead to a home bias. For instance, they might want to hold
a portfolio that is hedged against unanticipated changes in the purchasing
power of their currency, that has a return negatively correlated with the return
to their human capital, or that has a return correlated with the spot rate of
interest. One can think of many wtriables that affect the expected utility and lhe
portfolio selection of an investor. When investors hedge against unanticipaled
changes in state variables, limited diversification may be optimal even though
markets are fully integrated internationally. However, hedging against state
variables often seems to imply that investors should hold disproportionately
more foreign securities than domestic securities. For instance, it would seem
more likely that foreign securities have returns negatively correlated with the
returns to human capital and with the purchasing power of the investor's
domestic currency. Cooper and Kaplanis (Jk994) show, for instance, that hedging
purchasing power risks cannot explain the home bias. Using a fully specified
general equilibrium model, Uppal (1993) finds that there is a home bias only if
risk aversion is low. If investors are more risk averse than investors with
logarithmic utility as one would expect, hi,,; model predicts a reverse home bias.
Finally, Baxter and Jermann (1993) and Baxter et al. (1994) argue that taki:ag
into account the return to human capital should lead US investors to be short in
the US market portfolio of traded securities because of its high correlation with
the return to human capital. The implicatJions of hedging demands at the firm
level cannot be made precise without positing what investors hedge against.
Unfortunately, we are not aware of a well-specified hypothesis for state-variable
risks predicting that investors hold portfoliios with a substantial home bias. We
therefore provide no evidence on hedging demands as a possible explanation for
the home bias.

3. The data

In Japan, shares are registered. Firms report foreign ownership and this
information is available from annual reports and stock guides. Foreign owner-
ship includes all shares held by non-residents irrespective of where they are
located. It includes institutional as well as individual ownership by non-reside:at
investors. It also includes ADRs and all ADRs are counted as foreign ownership.
The ownership data are available from the Pacific-Basin Capital Market Re-
search Center (PACAP) files. The file includes all firms in existence as of the
beginning of 1991. In this paper, we use the data reported on these files from
1975 to 1991. For each firm, the file provides foreign ownership as of the end of
the fiscal year (which is March 31 for many Japanese firms). Table 1 provides
a summary of our data. In the second column, we show for each year the number
of firms for which foreign ownership data are available and the number of firms
 J.-K. Kung, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28 9

Table 1
Equally and value-weighted foreign ownership for nonfinancial Japanese firms by year.

Sample size Equally-weighted

Year (missing)" [Standard deviation] Value-weighted

1975 868 2.46% 4.64%

(385) [7.71]
1976 988 2.34 4.47
(267} [7.22]
1977 1,014 2.36 4.02
t.250) [7.10]
1978 1,043 2.35 3.39
{230) [6.86]
1979 1,089 2.30 3.25
( 191 ) [6.67]
1980 1,106 2.73 4.80
(187) [7.23]
1981 1,113 4.08 7.98
(2041 [8.44]
1982 1,118 4.45 8.80
{235) [8.60]
1983 1,144 5.07 10.46
(251) [8.96]
1984 1,191 5.76 11.31
(217) [8.96]
1985 1,308 5.15 9.97
(109} [8.28]
1986 1,298 5.17 8.97
(93J [7.96]
1987 1,268 4.25 6.88
(77) [7.25]
1988 1,309 3.77 4.50
(52) [6.74]
1989 1,341 3.71 5.02
(82) [6.11]
1990 1,426 3.99 4.76
(29) [6.38]
1991 1,452 4.02 5.59
(13) [6.5/)3
Average of annual values 3.76 6.40
Standard deviation of annual values 1.16 2.64

aMissing means the number of observati,ons for which foreign ownership or market value of equity is
not available at the end of the fiscal year.
10 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

for which foreign ownership or market value of equity data are missing from the
files. It is immediately apparent that the proportion of firms for which data are
missing falls steadily throughout our sample period. In the last sample year,
1991, we have data for 1,452 firms and 13 firms have missing data. In contrast, in
1975 (the first sample year), we have data for 868 firms and missing data for 385
firms.
To obtain summary statistics of foreign ownership, we could proceed in ,at
least two different ways. First, we could compute the percentage of shares owned
by foreign investors for each firm and average this percentage across firms. This
number is given for each year in the third column. In brackets, we also provide
the cross-sectional standard deviation of the percentage of shares owned by
foreign investors in each firm. The equally weighted measure of foreign owner-
ship is quite small. It never exceeds 6%. It exhibits a hump-shaped pattern:
foreign ownership first increases and then falls, reaching a peak in 1984. This
hump-shaped pattern is also documented by French and Poterba (19901 who
use aggregated Tokyo Stock Exchange data. They argue that such a pattern is
especially puzzling in light of the fact that barriers to international investment
decrease throughout the 1980s for investors wanting to invest in Japan. Second,
we could add up the market capitalization of all the firms for which we have dal:a
and compute the market value of the shares held by foreign investors as
a percentage of this market capitalization. We provide this measure in the last
column of Table 1. The value-weighted measure of foreign ownership is always
larger than the equally weighted measure and always by at least 50%. This
difference reflects an important characteristic of the portfolios held by foreign
investors that we will discuss throughoul the paper: foreign investors have
disproportionately more shares of large firms in their portfolios.
The firms with missing data are not included in the estimates of foreign
investment reported in Table 1. If these firms have more foreign investment than
the firms used in that table, then we are underestimating foreign investment. We
find, however, that including the firms for which the market value of equity is
missing but foreign ownership data are available in our equally weighted
measure of foreign ownership does not change our conclusions. In any case, few
firms have missing data in the last years of our sample period. Another issue
with our data is that there could be reporting errors or biases that lead some
foreign investors to be counted as domestic investors and some domestic
investors could be counted as foreign investors because they use a foreign
vehicle to invest in Japan. We have no clear way of assessing the importance of
these biases. Finally, Japanese firms issued large amounts of convertible debt
and debt with warrants offshore. Kang et al. (1995) provide an analysis of
Japanese offshore issues. It could be that the obstacles to holding offshore debt
were smaller for foreign investors than the obstacles to holding domestic shares,
so that foreign investors substituted offshore equity-linked debt for domestic
shares. Since the offshore issues are largely concentrated in the late 1980s, this
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28 11

Percent
50

40

30

20

10

1975 1977 1979 1981 1983 1985 1987 1989 1991

Year
l=-Japan market portfolio ==Foreign portfolio I

Fig. 1. Market value of Japanese shares held by foreign investors ;and of the Japanese market
portfolio in percent of the world market portfolio.
The world market portfolio is the one of Morgan Stanley Capital International measured at the end
of March. The value of Japanese shares held by foreign investors and the Japanese market portfolio
are measured at the same time.

substitution could play a role in the decrease in foreign ownership in the late
1980s documented in Table 1.
Fig. 1 shows the extent of the home bias over our sample period. If investors
hold the world market portfolio, the weight of Japan in their portfolio is equal to
the weight of Japan in the world market portfolio. The weight of Japan in the
world market portfolio is plotted in Fig. 1 using the Morgan Stanley Capital
International world market portfolio. 4 We also plot the value of the portfolio of
Japanese stocks held by foreign investors as a percentage of the value of the
world market portfolio. This percentage corresponds to the fraction of the
Japanese market portfolio held by foreign investors mulliplied by the fraction of
the world market portfolio represented by Japanese stocks ( x 100). If foreign
investors held the world market portfolio, their portfolio of Japanese stocks
would constitute the same percentage of the world market portfolio as the
market portfolio of Japanese stocks. Not surprisingly, foreign investors always
hold disproportionately less of the Japanese market portfolio. The difference
between the two weights represents the shortfall in holdings of Japanese stocks
by foreign investors relative to the case where they would hold the world market
portfolio. Since the portfolio of Japanese stock of foreign investors is always
a small fraction of the world market portfolio, the shortfall is highly correlated

"~We are grateful to Campbell Harvey for helping us find this data.
J.-K. Kang, R.M. Stulz/'Journal of Financial Economics 46 (1997) 3 28

• O c q ~ ' ~ , ~ ' ~ ' , ~ , ~ v ' ~ t ' - - ~ "

E
Z

& =
• ~ ~ ~ ~ ~
e~ ~:~

e.

x.D

e.,
"E

"6
o~ t ~ ~ " ~ ~ ~ ~ ~ o~

E ~

c-

e-

L~

e &

.~_ e-,
~

_~..~ ~
f~
 J.-K. Kang, R.M. Stulz /'Journal of Financial ~conomics 416 (1997) 3-28 13

~ o o

I ~ I ~ I ~ I ~ i ~ I ~ I ~ I ~ I V

I ~- ,-~ ~ ~'~ ~ 0 ('~1 ~ ~

I ~-" I ~"~-~ I ~"~" I ~'~-" I I I I V

~ ~-~ ~-o ~= ~.= ~-- ~o ~= ~ . = o°.

t -~ ~ ~ I ---, - - - ---, i ~- t v

. . . . . . . . "I. . . . . . . . ~.~.

to ~ q . . . . Z - iv

i.d
i ~ i ~ I . . . . . .

~ ,7., ~ o'~ ,~ .,~

14 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

with the share of the Japanese market portfolio in the world market portfolio
and keeps increasing with that share during the 1980s even though Japanese
capital markets become less regulated.
Table 2 provides a different perspective on foreign ownership, lnstead of
looking at the whole market, we look at separate industries. For each industry,
we report the difference between the industry's value-weighted foreign owner-
ship and the industry's weight in the Japanese market portfolio. The numbers
are in percentage terms, so that 1% means that foreign investors invest 1% more
of their Japanese portfolio in an industry than they would if their investment
weights were those of the Japanese market portfolio. There are striking patterns
of foreign ownership across industries. Throughout our sample period, foreign
investors hold disproportionately more of the manufacturing sector every year,
less of the utilities sector every year, and more of the services sector every year
but one. Surprisingly, they hold disproportionately less of the real estate sector
every year for the first eight years of our sample and more every year afterwards.
A similar pattern holds for construction. The deviations for manufacturing are
extremely large throughout the sample. On average, foreign investors invest
13.70% more of their Japanese portfolio in manufacturing than in the market
portfolio. Because the deviations are fairly stable over time, even small average
deviations are significantly different from zero. The only average deviation that
is not significantly different from zero is for real estate. A surprising result is the
dramatic increase in the bias against the transportation industry starting in
1988. The capitalization of that industry increases strongly that year, but since
foreign investors invest disproportionately less in that industry, they do not
participate in that capitalization increase as much as the market portfolio.

4. Firm characteristics and foreign ownership

In this section, we use multivariate regressions to explore how ownership cf

Japanese shares by non-Japanese investors is related to firm characteristics. We
pursue two different approaches. In the first approach, we estimate regressions
each year. The advantage of this approach is that every year we can include in
our sample all firms for which we have data. There is therefore no requirement
that we have complete information on each firm every year. The disadvantage cf
these regressions is that they make no use of the time-series information. In the
second approach, we use times-series cross-section regressions. These regressions
require a fixed sample so that they suffer from sample selection bias. However,
they take full account of the time-series information. As we will show, the results
with the second approach are stronger, bm they are fully consistent with the
results from the first approach. We use the tollowing firm characteristics:
(a) Leverage. Leverage is measured as the ratio of total liabilities to total
assets at the end of the fiscal year. Leverage is fairly high for Japanese firms
 .~-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28 15

compared to firms in other countries for much of our sample period. If foreign
investors believe that the leverage of Western firms is more appropriate, one
would expect them to underinvest in the Japanese firms with the highest leverage.
(b) Current ratio. We use the current ratio, defined as the ratio of current
assets to current liabilities at the end of the fiscal year, as a measure of short-run
financial health of a firm.
(c) R e t u r n on assets. Return on assets is defined as net income divided by total
assets as of the end of the fiscal year.
(d) Beta. Beta is the market model beta estimated using daily returns for the
previous calendar year (e.g., if foreign ownership is measured in March 1987, the
end of the fiscal year, then the corresponding beta is calculated using 1986 daily
returns). The market portfolio is; the Japanese equally weighted portfolio from
the P A C A P files. Models of barriers to international investment that treat these
barriers as proportional taxes generally conclude that investors facing such
barriers when they invest abroad hold disproportionately more foreign high
beta stocks.
(e) Residual variance. Residual variance is the variance of the market model
error estimated using daily returns for the previous calendar year.
(t) E x c e s s return. Excess return is measured as the return net of the equally-
weighted P A C A P portfolio return. It is measured as the cumulative compound
excess return using the 12 monthly returns preceding the end of the fiscal year
(e.g., for foreign ownership measured in March 1987, the excess return is
calculated using monthly returns from April 1986 to March 1987). Including this
variable makes it possible to ewtluate whether foreign investors are contrarian
or extrapolative.
(g) M a r k e t value. This is the market value of the firm's; shares at the end of the
fiscal year. Size could play a role in portfolio allocations. First, it is often argued
that more information is available about large firms, so that information
asymmetries between Japanese and non-Japanese investors might be less impor-
tant for such firms. Second, transaction costs are lower for such firms, so that if
barriers to international investment include higher transaction costs for foreign
investors, it is possible that they are lower for such firms. Third, foreign investors
are more likely to know about large firms.
(h) B o o k - t o - m a r k e t . The book-to-market ratio is measured as the book value
of equity divided by the market value of equity as of the end of the fiscal year.
We present the first set of results in Table 3. For these results, we proceed as
follows. Each year, we estimate a cross-sectional regression of foreign ownership
in a firm on eight explanatory variables. Note that if foreign investors hold the
market portfolio of the domestic country, foreign ownership is the same in each
firm and the coefficients of our cross-sectional multivariate regression should be
insignificantly different from zero. Hence, these coefficients show how foreign
investors depart from the market portfolio in their holdings. Rather than
reproduce the coefficient estimates for each year, we average them first for the
16 J.-K. Kang, R.M. Stulz//Journal of Financial Economics 46 (1997) 3-28

whole sample period and then for two subsamples. We provide three different
measures to help assess the statistical significance of the results: the average of
the t-statistics for a coefficient, the number of t-statistics for a coeffici~;nt
significantly different fi'om zero at least at the 0.10 level, and the p-value of
a t-test for the average t-statistic.
The results for the first column of Table 3 aggregate the coefficient estimal:es
for 16 regressions. The only explanatory variable that has a significant coeffic-
ient each year is firm size. Each year, foreign investors invest more in large firms
controlling for the seven other firm characteristics. The coefficient on leverage is
significantly negative in 12 years out of 16, indicating that foreign investors
prefer firms with low leverage. There is a strong book-to-market effect in the first
subperiod, but this effect weakens substantially in the second subperiod. In
contrast, the return-on-assets effect is strong in the second-half of the sample,
but much less so in the first half. The other variables sometimes have significant
coefficients, but they do not provide convi~acing results. The result that domin-
ates Table 3 is the importance of firm size in the investment decisions of foreign
investors. One might be concerned that this reflects a reporting bias, in that the
smaller firms might be more likely to have missing observations. However, if
that were the case, one would expect the effect should falter in the second-half of
the sample and it does not. An alternative way to regress foreign ownership on
our explanatory variables is to pool years together and allow firms to enter in
the panel as they become available. We do so allowing for dummy variables fi~r
each year. This alternative approach assumes that the coefficients are consta~lt
through time, whereas the results in Table 3 allow the coefficients to change
through time. Nevertheless, the results are similar except that the book-to-
market ratio and beta are much more significant. This pooling approach does not
account for correlation in the residuals, so that the t-statistics might be inflated.
Table 4 provides regressions that use both cross-sectional and time-series
data. In that table, we estimate random-effect models using the Fuller-Batte~,;e
(1974) method. This method divides the error term for a firm at year t into thr~;e
components: an error for firm i across years, an error for firm i in year t and an
error for year t common across firms. The variance components are estimated
first by the fitting-of-constants method and the regression parameters are
estimated using generalized least squares. T,~ implement this approach, we have
to use a fixed panel, however. Since we are interested in deviations of the
portfolio held by foreign investors from the market portfolio for Japan, we use a.s
our dependent variable foreign ownership tbr a firm in a given year minus the
equally weighted foreign ownership for that year. The dependent variable
therefore measures how the foreign ownership in a firm differs from what :it
would be if foreign investors had acquired the same fraction of each firm. The
regressions in Table 4 show again a strong effect of firm size on ownership.
Leverage and book-to-market have significantly negative coefficients for the
whole sample regression and the two subsample regressions and ROA has
 J.-K. Kang, R.M. Stulz/Journal o f Financial Economics .¢6 (1997) 3 28 17

Table 3
Regression estimates of foreign ownership on explanatory variables. Regression coefficients are the
time-series average from year-by-year regressions for the 1976 1991 period? Average t-statistics are
in parentheses; the second n u m b e r in parentheses is the p-value for a t-test that the average t-statistic
is zero. N u m b e r s in brackets are those of coefficients that are significantly positive at the 0.10 level
and those of coefficients that are signific, antly negative at the 0.10 level, respectively.

Full period: Subperiod: Subperiod:

Variables 1976-199l 1976 1983 1984-1991

Intercept - 0.0635 - 0.0370 - 0.0899

( - 2.19; < 0.01) ( - 1.09; 0.01) ( - 3.29; < 0.01)
[0,10] [0,?.] [0,8]
Leverage - 0.0397 - 0.0452 - 0.0342
( - 2.03; < 0.01) ( - 2.03; < 0.01} ( - 2.03; < 0.01)
[0,12] [0,6] [0,6]
C u r r e n t ratio 0.0017 0.0.310 0.0024
10.48: 0.04) (0.23; 0.44) (0.74; 0.06)
[1,0] [0,0] [l,0]
ROA 0.1495 0.1683 0.1308
(l.72; < 0.01) (1.69; < 0.01) (1.75; < 0.01)
[8,0] [3,0] [5,0]
Beta 0.0025 0.0006 0.0044
(0.61; 0.24) (0.14; 0.80) (1.07; 0.24)
[4,1] [1,0] [3,1]
Residual variance 6.7279 12.1359 1.3199
(0.76; 0.08) (IAI; 0.08) (0.10; 0.73)
[4,0] [3,0] El,0]
Excess return 0.0036 0.0049 0.0022
(0.38; 0.55) (0.45; 0.65) (0.31; 0.69)
[4,3] [2,1 ] [2,2]
Book-to-market - 0.0246 - 0.0390 0.0101
( -- 2.10; < 0.01) ( -- 3.28; < 0.01) ( -- 0.93; 0.18)
[1,10] [0.7] [1,3]
Log (MV) 0.0127 0.0114 0.0140
(6.57; < 0.01) (5.22; < 0.01) (7.92; < 0.01)
[16,0] [8,0] [8,0]
Average sample size 1,122 993 1,251
Adjusted R 2 (°/o) 11.626 10.413 12.840

aFirms with an extreme R O A ( R O A larger than 1 or smaller than - 1) a n d firms with negative b o o k
equity are deleted.
18 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

Table 4
Time series cross-section regression estimates of foreign ownership on explanatory variables. The
Fuller-Battese method is used to estimate the model. ,Only 519 firms with complete data during the
1975 1991 period are used (t-statistics are in parentheses).

Full period: Subperiod: Subperiod:

Variables 1976-1991 1976 1983 1984 1991

Intercept -- 0.1883 - 0.2022 - 0.1092

[ -- 9.33) - 7.4O) - 4.27)
Leverage -- 0.0284 - 0.0795 - 0.0278
( - 3.74) - 6.43) - 2.58)
Current ratio 0.0010 0.0022 - (/.0012
(1.379) (0.95) - 1.541
ROA 0.1121 0.0596 (I.1730
(4.72) (1.951 (5.47)
Beta 0.0031 - 0.0008 0.0070
(3.53) - 0.77) I5.55~
Residual variance - 2.5213 2.1579 - 11.4160
( - 1.43) (l.18} (-4.83)
Excess return 0.0059 0.0020 0.0072
(4.71) 11.271 (4.76)
Book-to-market 0.0211 - 0.0138 - 0.0490
[ - 5.66) - 2.93) ( - 8.36)
Log (MV) 0.020l 0.0261 0.0132
(15.03) {13.47) (7.38)
Root MSE 0.03573 0.02961 0.03175

a s i g n i f i c a n t p o s i t i v e coefficient for the t h r e e regressions. N o n e o f the o t h e r

v a r i a b l e s h a v e coefficients t h a t a r e significant in b o t h s u b p e r i o d s . In c o n t r a s t to
T a b l e 3, all coefficients in t h e s e c o n d s u b p e r i o d are s i g n i f i c a n t e x c e p t t h e o n e fc,r
the c u r r e n t ratio. In the s e c o n d s u b p e r i o d , b e t a a n d the excess r e t u r n h a v e
significant p o s i t i v e coefficients a n d the r e s i d u a l v a r i a n c e has a significant neg~,-
t i r e coefficient.

5. An investigation of the size bias

A useful w a y to u n d e r s t a n d the m a g n i t u d e o f the size bias is to l o o k at the

c o r r e l a t i o n b e t w e e n firm m a r k e t v a l u e a n d f o r e i g n o w n e r s h i p . W e c o m p u t e t h e
S p e a r m a n r a n k c o r r e l a t i o n b e t w e e n f o r e i g n o w n e r s h i p a n d t h e l o g of the m a r k e t
v a l u e e a c h year. T h e a v e r a g e S p e a r m a n r a n k c o r r e l a t i o n is 0.455; it is signifi-
c a n t l y p o s i t i v e e v e r y y e a r at the 0.01 level. W h y this e x t r a o r d i n a r y size bias? Size
c o u l d p r o x y for several v a r i a b l e s t h a t m i g h t affect f o r e i g n o w n e r s h i p . In this
section, we try to assess this p r o x y effect.
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3-28 19

5.1. Size and the Merton effect

Merton (1987) argues that investors hold shares in firms with which they are
familiar and that investors are more likely to be familiar with large firms.
Falkenstein (1996) shows that mutual funds hold more shares in firms that have
a lot of news stories associated with them. A proxy for how well known a firm is
to foreign investors is the extent to which a firm exports. For a subset of firms
(almost all manufacturing firms), we have data made available by Daiwa
Securities Co. that provide the ratio of exports to sales for the parent company.
These data suffer from the fact that they do not include information for
unconsolidated foreign subsidiaries. On average, the data are available for 678
firms per year. The average ratio of exports to sales is 0.1612 per year. Comput-
ing the correlation between foreign ownership and the exports ratio, we find an
average correlation of 0.18. This correlation is positive every year and significant
at the 0.01 level for 16 years and at the 0.05 level for one additional year. It
appears from these correlations that foreign investors invest more in firms that
have a higher export ratio. However, larger firms are likely to have higher export
ratios, so that the correlations might just reflect the correlation between size and
foreign ownership documented earlier.
To separate the effect of the export ratio and the effect of size, we present in
Table 5 foreign ownership for size quintiles and foreign trade quintiles. Ignoring
size, foreign ownership increases monotonically with the ratio of exports to
sales, going from 3.69% for the lowest export ratio quintile to 5.71% for the
largest. The impact of size is much larger, however, since foreign ownership is,
on average, 1.80% for the smallest quintile and 7.66% for the largest quintile.
Among the largest firms, the ratio of exports to sales is uninformative: foreign
ownership is 8.16%, on average, for the lowest quintile of the exports ratio and
8.50% for the largest. In contrast, for small firms, foreign trade seems to affect
foreign ownership. In particular, for the smallest size quintile, the firms in the
smallest export ratio quintile have average ownership of 1.03% and in the
largest export ratio quintile average ownership is 2.82%. The average difference
between these two quintiles is significant with a t-statistic of 4.67. The difference
between the high and low exporters for the next size quintile is significant
also, but the differences for the ,other size quintiles are not significant. There is
therefore some evidence that foreign activities, matter for foreign ownership if
a firm is small but not if it is large. Note, however, that our measure of the
ratio of exports to sales is likely to be much less informative for large firms
since many of these have foreign activities through unconsolidated subsidiaries.
In constructing Table 5, we ignore firms for which we have no export data.
An alternative approach is to assume that these firms have no exports. If
Table 5 were constructed that way, the qualitative results would be similar,
but the export ratio quintiles appear somewhat more related to foreign
ownership.
20 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3-28

Table 5
Mean and median foreign ownership(%) by portfoliostormed on the market value of equity and th~,'n
the exports-to-sales ratio during the 1975-1991 period. Each year, the firms for which data are
available on the exports-to-sales ratio are divided into size quintiles.Each quintile is then divided inLo
five quintiles based on the exports-to-sales ratio. The cells in the table provide the time-series me~tn
(median) of the yearly means (medians).We also provide the average difference between the largest
and smallest quintiles with the t-statistic computed from the time-series of the yearly differences.

Size quintiles
Exports/Sales
ratio Smallest 2 3 4 Largest All

Smallest (1) 1.03 2.63 3.39 4.87 8.16 3.69

(0.96) (2.85) t3.36) (5.04) (9.62) (3.84)
2 1.41 2.86 3.95 5.94 6.69 3.74
(1.46) (3.03) (3.22) (5.79) (6.04) (3.90)
3 2.26 2.44 3.99 5.09 7.73 4.10
(2.55) (2.25) (4.34) (4.15) (5.92) (3.81)
4 1.46 3.21 3.47 6.44 7.18 4.46
(1.45) (3.11) (3.86) (5.57) (6.08) (4.25)
Largest (5) 2.82 4.15 4.17 4.72 8.50 5.71
(2.78) (3.62) (4.30) (4.26) (8.68) (5.68)
(5) - (1) 1.79 1,52 0.78 - 0.15 0.34 2.02
[t-statistic] [4.67] [2,95] [1.18] [ - 0.17] [0.33] [3.44~[
All 1.80 3,05 3.79 5.4l 7.66
(1.78t (3.07) (4.10) (5.15) (6.83)

An alternative a p p r o a c h to investigate 'whether the export ratio plays al~

i m p o r t a n t role in explaining the coefficient on size in our cross-sectional regres-
sions is to reestimate these regressions with the export ratio as a n a d d i t i o n a l
e x p l a n a t o r y variable. We reestimated the regressions from Tables 3 a n d 4 o n the
s u b s a m p l e of 263 c o m p a n i e s for which we have export data every year. The
export ratio has an insignificant coefficient in the foreign ownership regressions
for the whole sample a n d the first-half of the sample. It has a positive coefficient
of 0.04 with a t-statistic of 2.39 for the second-half of the sample (1984 1991), in
the time-series cross-section regression a n d an average coefficient of 0.039 with
an average t-statistic of 2.34 for the cross-section regressions. A d d i n g the export
ratio has little i m p a c t on the size coefficient. C o n s e q u e n t l y , while the export
ratio seems to be a d e t e r m i n a n t of lbreign ownership, it c a n n o t explain the size
coefficient in our regressions.

5.2. Size and liquidity

It is less costly to build positions in shares that are more liquid. I f a share lacks
liquidity, those who k n o w the m a r k e t well have an advantage. They can time
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3-28 21

Table 6
Mean and median foreign ownership 1%t by portfolios formed on the market value of equity and
then the turnover ratio (defined as volume divided by the number of shares outstanding) during the
1975-1991 period. Each year, the firms for which data are a,~ailable on '~he turnover ratio are divided
into size quintiles. Each quintile is then divided into five quintiles based on the turnover ratio. The
cells in the table provide the time-series mean lmedian) of the yearly means (medians). We also show
the average difference between the largest and smallest quintiles with the t-statistic obtained from
the time-series of annual differences.

Size quintiles

Turnover ratio Smallest 2 3 4 Largest All

Smallest (1) 0.83 2.06 2.38 3.90 4.80 2.71

(0.86) (2.1)2) (2.36) (3.82) (4.92) (2.64)
2 1.05 2.22 3.27 5.50 6.48 3.33
(0.98) (2.27) (3.21) (5.49) (6.21) t3.37)
3 1.27 2.32 4.36 5.57 7.85 4.10
(t.33) (2.37) (4.62) (5.95) (6.64) (3.49)
4 1.44 2.72 3.48 4.90 8.45 4.49
[ 1.40) (2.85) (3.44) (5.17) (7.70) (4.58)
Largest (5) 1.44 2.:58 3.08 4.82 7.27 4.16
(l.40t (2.79~ (3.21) (5.19) (7.12) (4.27)
(5) - (1) 0.61 0.:52 0.70 0.92 2.47 1.45
It-statistic] [2.43] [1.47] [l.22] [1.44.] [2.78] [2.92]
All 1.21 2.38 3.31 4.94 6.97
(1.10) (2.46) (3.59) (4.97') (6.47}

their purchases to take advantage of their information about the supply of

shares. In such a setting, foreign investors might well prefer large firms because
the market for their shares is more liquid. Further, as argued in Section 2,
concerns about political risk might also induce investors to hold liquid shares
since they can exit their positions at lower cost. We do not have a measure of
liquidity that we can use. However, shares with higher turnover (defined as
volume divided by the number of shares outstanding) should be more liquid. We
create an annual turnover measure by using monthly volume to obtain a daily
average turnover measure for a month. We then average the daily average
turnover measures across the year.
Table 6 makes it possible to address the issue of whether size proxies for
liquidity. We rank shares by size and then divide each size quintile into five
turnover quintiles. Turnover se,ems to be related to foreign ownership. For
a given size quintile, the lowest turnover quintile always; has substantially lower
ownership. The turnover effect does not seem to be present for the other
turnover quintiles. The average differences between the highest and lowest
turnover quintiles are significant for the smallest and largest size quintiles as well
22 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

as for the whole sample. We also reestimate Tables 3 and 4 using the turnover
variable as an additional independent variable. The coefficient on turnover iis
positive and significant in the cross-section regressions only in the last two years
of the sample. In the time-series cross-section regressions, the coefficient on

Table 7
Event study of foreign ov, nership when ADRs become available for a firm. Year 0 is the fiscal year
during which ADRs become available. When a year is before 1975, we do not have available data.
A matching firm is chosen for each ADR firm using market values in year 0.

ADR firms Matching firms Mean ownership difference

Mean ownership Mean difference (t-statistic for difference)
[median] [median ] [median difference]
Sample size Sample size [Wilcoxon-z for difference',
(%l (%) (%)

Year - 2 11.61 2.42 9.19

(1.96)*
[11.201 [1.88] [9.32]
6 6 ',1.0<
Year - l 8.37 4.00 4.37
(1.75)*
[9.57] [2.93] [6.64]
lO 10 {1.321
Year 0 10.21 5.12 5.09
(1.69)
[10.17] [3.56] [6.61]
11 11 1.581
Year + 1 7.98 5.51 2.47
(1.12)
[6.92] [3.292 [6.61]
14 14 ~t.o81
Year + 2 6.39 3.57 2.82
(1.85)*
[5.321] [2.66] [2.66]
14 14 {1.681"
Year -r 3 5.68 2.63 3.05
(2.92)**
[_4.01] [ 1.74] [2.27]
26 26 12.641 **
Year + 4 7.63 3.80 3.83
(2.74)**
[7.381] [2.95] [4.43]
27 27 [2.511 **

*Significant at the 0.10 level.

**Significant at the 0.01 level.
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28 23

turnover has an unexpected negative sign and is significant for the whole sample
and for the second-half of the sample. Again, however, the additional explana-
tory variable has little effect on the size coefficient.

5.3. The barriers to international investment effect

Could it be that explicit barriers to international investment are smaller for

large firms? Large firms are more likely to have A D R programs. Such programs
could lower the pecuniary barrie, rs to international inw~stment. In Table 7, we
provide the results of an event study of foreign ownership. The event year is the
year that an A D R program starts for a firm. Unfortunately, many ADR
programs for the firms in the sample started before 1975, so that we have no
ownership data for m a n y firms in the event year. Each year we use all the firms
for which we have data. We find that the A D R firms have more foreign
ownership than firms of similar size at year 0. However, somewhat surprisingly,
there is no evidence in the table that the ADR program itself increases foreign
ownership. It seems rather that firms with more foreign ownership choose to
have ADR programs. Firms with ADR programs are mainly firms in manufac-
turing and generally are well known in the US for their brand goods. It may well
be, therefore, that these firms have ADRs to reduce the costs of share ownership
for existing shareholders and that they have more shareholders abroad because
they are well known.

6. The investment performance of foreign investors

Barriers to international investment that take the form of a deadweight cost

imply that the portfolio of domestic securities held by foreign investors should
have a higher expected return than the portfolio held by domestic investors
before taking into account the deadweight cost. Table 8 provides yearly average
excess returns of foreign investors over the P A C A P monthly value-weighted
return and associated t-statistics. To construct the returns of foreign investors,
we proceed as follows. For year t, we use ownership intbrmation at the end of
fiscal year t - 1. We use market value information as of the end of June of year
t to construct a portfolio that mimics holdings of foreign investors and then
compute the monthly return on a portfolio with these weights for the next 12
months. Note that the excess returns for foreign investors are obtained by taking
the difference of two yen returns. They therefore have the interpretation of
excess returns for foreign investors irrespective of their currency if the optimal
currency hedge for Japanese securities is to go short an amount of yen equal to
the price of the security. We investigate the correlation between yen returns on
the portfolio held by foreign investors and the yen/dollar exchange rate and find
the correlation to be economically trivial and statistically insignificant. The
24 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28

Table 8
Time-series average excess returns earned by foreign investors during the period 1976 1991 period.
Excess return is the difference between value-weighted monthly returns of foreign investors (where
the weight is the market value of each Japanese company held by foreign investors divided by the
total market value of Japanese shares held by foreign investors) and value-weighted PACAP
monthly market returns. Foreign ownership is measured in fiscal year t - 1 and market value is
measured in June of year ~. These values are matched with returns for the months from July of year
t to June of year t + 1.

Average monthly
Period excess return r-statistics

76/07 -- 77.,,'06 -- 0.0699% - 0.09

77/07 78/'06 -- 0.3734 - 0.63
78/07- 79./'06 0.7072 0.97
79/07 80/06 -- 0.1564 - 0.30
80/'07 - 81/06 0.4017 0.83
81,,"07 82/06 -- 0.6305 -- 0.71
82/07 83/06 0.9495 1.74
83/07 - 84/06 -- 0.7257 -- 1.36
84/07 85/06 - 1.2831 -- 1.32
85/07 86/06 -- 0.2069 -- 0.15
86/07 87/06 -- 0.6993 -- 0.32
87/07 88/06 0.4136 0.59
88/07 89/06 0.7365 2.40*
89/07 90/06 0.5461 1.03
90/07 91/06 0.2565 0.46
91/07 91/12 - 0.3085 -- 0.53
76/07 84/06 0.0128 0.06
84/07 - 91/12 - 0.0520 - 0.13
76/07 91/12 - 0.0190 - 0.08

*Significant at the 0.05 level.

p r o c e d u r e we use a l l o w s us t o o b t a i n r e t u r n s f o r 16 y e a r s . O u t o f t h e s e 16 years.,
foreign investors underperform the value-weighted PACAP index portfolio nine
times. T h e i r a v e r a g e e x c e s s r e t u r n r e l a t i v e t o t h e v a l u e - w e i g h t e d P A C A P is
n e g a t i v e o v e r t h e w h o l e s a m p l e p e r i o d a n d in t h e s e c o n d s u b s a m p l e p e r i o d ,
w h i c h is t h e p e r i o d f r o m 1984 t o 1991. T h e u n d e r p e r f o r m a n c e is n o t s i g n i f i c a n t ,
however.
Did foreign investors choose a portfolio with a greater expected return than
t h e m a r k e t p o r t f o l i o , as o n e w o u l d e x p e c t if t h e e x p l a n a t i o n for t h e h o m e b i a s is
a p r o p o r t i o n a l d e a d w e i g h t c o s t ? W e h a v e s e e n t h a t t h e i r p o r t f o l i o is w e i g h t e d
t o w a r d s l a r g e firms, w h i c h w o u l d c o r r e s p o n d t o a s m a l l e r e x p e c t e d e x c e s s
return. In addition, the beta of their portfolio with respect to the value-weighted
P A C A P p o r t f o l i o is less t h a n one. I n t e r e s t i n g l y , t h o u g h , t h e b e t a is m u c h h i g h e r
i n t h e f i r s t - h a l f o f t h e s a m p l e (1.23) t h a n i n t h e s e c o n d - h a l f o f t h e s a m p l e (0.84).
 J.-K. Kang, R.M. Stulz/Journal of Financial F,conomics ,I6 (1997) 3 28 25

Hence, investors did choose a higher-risk portfolio when barriers to interna-

tional investment were higher.
By choosing a portfolio weighted towards large firms, foreign investors did
not perform significantly worse than the market portfolio. They did, however,
hold a portfolio with a greater standard deviation than if they had held the
market portfolio. The standard deviation of the monthly return on the market
portfolio is 4.81% over our sample period and 5.38% fi~r the portfolio held by
foreign investors. (The p-value for the F-test that the variances are equal is
0.1287.) In other words, foreign investors had a portfolio with a higher volatility
than the market portfolio without any gain in expecled return. This means
that the portfolio they were holding was a poor proxy for the market portfolio.
Another indication of this is that the R 2 in a regression of the foreign
portfolio return on the market portfolio return is 69%. As explained
in Section 2, however, investors might have wanted a portfolio that correlates
in specific ways with state variables they care about. It is therefore possible
that the portfolio they held had tlhese correlations and that the market portfolio
did not.

7. Does foreign investment affect expected returns on domestic securities?

The evidence presented in this paper shows that foreign investors increase the
demand of some domestic securJities relative to other domestic securities. One
would expect that this differential demand affects expected returns. In this
section, we investigate the relation between expected returns on domestic
securities and the holdings of foreign investors. We proceed in the same way as
Fama and French (1992) to find out whether foreign ownership is related to the
cross-sectional variation in expected returns. Each month, we regress the indi-
vidual stock returns of that month on foreign ownership. We use our previous
timing convention in that we use the end of fiscal year t - 1 ownership data for
12 months starting with July of year t.
Table 9 provides the average slopes of the monthly regressions with
t-statistics obtained from their time-series standard deviation. The average slope
for the whole sample is negative but not signiticant. We also give the average
slope obtained from regressions ,of returns on the log of market value. There is
a size effect throughout our sample and it dominates the foreign ownership
effect. When we divide the sample in two subperiods, we find that the average
slope for foreign ownership is insignificantly positive for the first subperiod and
significantly negative for the second subperiod. There is no size effect in the first
subperiod but there is a size effect in the second subperiLod. When we allow for
both an ownership effect and a size effect, the average coefficient on foreign
ownership is not significant in the second subperiod and its absolute value falls
by almost two-thirds. Our evidence shows that there are some traces of foreign
26 J.-K. Kang, R.M. Stulz/'Journal of Financial Economics 46 (1997) 3 28

Table 9
R e g r e s s i o n e s t i m a t e s o f m o n t h l y r e t u r n s o n foreign o w n e r s h i p a n d m a r k e t v a l u e o f equity. Regres-
sion coefficients a r e the time-series a v e r a g e f r o m m o n t h - b y - m o n t h r e g r e s s i o n s f r o m J u l y 1976 to
D e c e m b e r 1991, a n d t-statistics a r e the a v e r a g e coefficient d i v i d e d b y its time-series s t a n d a r d error.
F o r e i g n o w n e r s h i p is m e a s u r e d in fiscal y e a r t - 1 a n d m a r k e t v a l u e is m e a s u r e d in J u n e o f y e a r t.
T h e s e values a r e m a t c h e d w i t h r e t u r n s for the m o n t h s f r o m J u l y of y e a r t to J u n e of y e a r t + 1.
A v e r a g e t-statistics a r e in p a r e n t h e s e s .

Model Foreign ownership L o g ( m a r k e t value)

J u l y 1 9 7 6 - D e c e m b e r 1991
( S a m p l e size = 186~
(1) - 0.0132
( - 1.48}
(2) -- 0.0025
- 2.58)
(3) - 0.0010 - 0.0025
( - 0.121 - 2.49)

J u l y 1 9 7 6 - J u n e 1984
( S a m p l e size = 96)
(4) 0.0002
~0.021
~51 -- 0.0019
- 1.471
(6) (I.0084 -- 0.0020
( 1.071 - -1.581

J u n e 1 9 8 4 - D e c e m b e r 1991
( S a m p l e size = 90)
(7) - 0.02758
( - 1.771
(8) - 0.0032
( - 2.141
(9) - 0.0111 - 0.0030
( -- 0.73) ( -- 1.921

ownership in returns. However, this effect seems weak and does not stand up
when we allow for a size effect. Although from the perspective of forecasting
expected returns it makes sense to allow for a size effect, the fact that foreign
investors seem to have greater demand for large-firm stocks may be a helpfill
clue in understanding why there is a size effect.

8. Conclusion

This paper provides evidence on the distribution of foreign share ownership in

Japan from 1975 to 1991. We confirm the existence of a substantial home bias
 J.-K. Kang, R.M. Stulz/Journal of Financial Economics 46 (1997) 3 28 27

u s i n g a different d a t a source. W e show, further, t h a t o w n e r s h i p b y foreign

i n v e s t o r s is c o n s i s t e n t l y a n d s t r o n g l y b i a s e d a g a i n s t s m a l l firms. T h e cost to
foreign i n v e s t o r s of o v e r i n v e s t i n g in large firms is t h a t their r e t u r n in J a p a n is
m o r e v o l a t i l e t h a n if t h e y held the J a p a n e s e m a r k e t portfolio. W e find that,
a l t h o u g h foreign i n v e s t o r s do n o l e a r n a greater r e t u r n t]han if they h a d held the
m a r k e t portfolio, the v o l a t i l i t y of their m o n t h l y r e t u r n is 5 . 3 8 % w h e r e a s the
v o l a t i l i t y of the r e t u r n of the ]market p o r t f o l i o is 4 . 8 1 % . M o d e l s t h a t are
c o n s i s t e n t with the existence of a h o m e bias are g e n e r a l l y i n c o n s i s t e n t with o u r
e v i d e n c e since these m o d e l s p r e d i c t t h a t foreign investor,; e i t h e r h o l d the m a r k e t
p o r t f o l i o of a c o u n t r y o r a p o r t f o l i o with d i s p r o p o r t i o n a t e l y m o r e securities t h a t
h a v e high expected excess returns. A m o d e l in w h i c h n o n r e s i d e n t i n v e s t o r s k n o w
m o r e a b o u t large firms t h a n small firms in the marke~Ls in w h i c h they i n v e s t
( o t h e r t h a n their h o m e m a r k e t c o u n t r y ) is o n e t h a t c a n p r o d u c e the cross-
s e c t i o n a l p a t t e r n s of o w n e r s h i p t h a t we d o c u m e n t .

References

Baxter, M., Jermann, U.J., 1993. The international diversification puzzle is worse than you think.
Working paper. University of Virginia, Charlottesville, VA.
Baxter, M., Jermann, U.J., King, R., 199,1. Nontraded goods, nontraded factors, and international
non-diversification. Working paper. University of Virginia, Charlottesville, VA.
Brennan, M.J., Cao, H.H., 1996. International portfolio investmenlE flows. Journal of Finance,
forthcoming.
Cooper, I.A., Kaplanis, E., 1994. What explains the home bias in portfolio investment. Review of
Financial Studies 7, 45--60.
Falkenstein, E.G., 1996. Preferences for stock characteristics as revealed by open-end mutual fund
portfolio holdings. Journal of Finance 51, 111-135.
Fama, E., French, K.R., 1992. The cross--section of expected stock returns. Journal of Finance 47,
427~,66.
Frankel, J.A., 1995. The Internationalization of Equity Markets. University of Chicago Press,
Chicago, IL.
French, K.R., Poterba, J.M., 1990. Japanese and U.S. cross-border common stock investments.
Journal of the Japanese and International Economies 4, 476 493.
French, K.R., Poterba, J.M., 199l. lnvestor diversification and internalfional equity markets. Ameri-
can Economic Review, Papers and Proceedings, 222 226.
Fuller, W.A., Battese, G.E., 1974. Estimation of linear models with crossed-error structure. Journal
of Econometrics 2, 67 78.
Kang, J.-K., Kim, Y.C., Park, K.-J., Stul2, R.M., 1995. An analysis of 1:hewealth effects of Japanese
offshore dollar-denominated convertible and warrant bond issues. Journal of Financial and
Quantitative Analysis, 257 270.
Low, A., 1993. Unpublished dissertation. University of California at Los Angeles, Los Angeles, CA.
Merton, R.C., 1987. A simple model of capital market equilibrium with imcomplete information.
Journal of Finance 42, 483 510.
Shiller, R.J., Kon-ya, F., Tsutsui, Y., 1990. Speculative behavior in the stock markets: Evidence from
the U.S. and Japan. Mimeo., Yale University.
Stulz, R.M., 1981a. On the effects of barriers to international investment. Journal of Finance 36,
923-934.
28 J.-K. Kang, R.M. Stu&/Journal of Financial Economics 46 (1997) 3 28

Stulz, R.M., Wasserfallen, W.+ 1995. Foreign equity investment restrictions, capital flight restrictions+
and shareholder wealth: theory and evidence. Review of Financial Studies 8, 1019 1057.
Tesar, L., Werner, I.M., 1995. Home bias and high turnover. Journal of International Money and
Finance 14, 467-493.
Uppal, R+, 1993. A general equilibrium model of international portfolio choices. Journal of Finance
48, 529-553.